Parameters: ----------- past_target_norm: torch.Tensor [batch, time steps, series] The historical data that will

Compute the loss function of the decoder. Parameters: ----------- encoded: Tensor [batch, series, time steps, embedd

Compute the loss function of the copula portion of the decoder. Parameters: ----------- hist_encoded: Tensor [batch,

Normalize the given values according to the historical data sent in the constructor. Parameters: ----------- value:

Create a large tensor and delete it. This operation occupies the GPU memory, so other processes cannot use the occupied memory. It is used to

Lightning hook that is called when a training epoch starts.

Generate num_samples future trajectories based on the provided context. context: DataFrame containing the most recent observations (

Prints the contents of a DictConfig as a tree structure using the Rich library. :param cfg: A DictConfig composed by Hydra. :param print_orde

Generate the given number of samples from the forecasted distribution. Parameters: ----------- num_samples: int

Generate the given number of samples from the forecasted copula. Parameters: ----------- num_samples: int

Set up which GPU we use for this process

Optional decorator that controls the failure behavior when executing the task function. This wrapper can be used to: - make sure loggers